1. |
- Forough, Javad, 1993-
(författare)
-
Machine learning for anomaly detection in edge clouds
- 2024
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Edge clouds have emerged as an essential architecture, revolutionizing data processing and analysis by bringing computational capabilities closer to data sources and end-users at the edge of the network. Anomaly detection is crucial in these settings to maintain the reliability and security of edge-based systems and applications despite limited computational resources. It plays a vital role in identifying unexpected patterns, which could indicate security threats or performance issues within the decentralized and real-time nature of edge cloud environments. For example, in critical edge applications like autonomous vehicles, augmented reality, and smart healthcare, anomaly detection ensures the consistent and secure operation of these systems, promptly detecting anomalies that might compromise safety, performance, or user experience. However, the adoption of anomaly detection within edge cloud environments poses numerous challenges.This thesis aims to contribute by addressing the problem of anomaly detection in edge cloud environments. Through a comprehensive exploration of anomaly detection methods, leveraging machine learning techniques and innovative approaches, this research aims to enhance the efficiency and accuracy of detecting anomalies in edge cloud environments. The proposed methods intend to overcome the challenges posed by resource limitations, the lack of labeled data specific to edge clouds, and the need for accurate detection of anomalies. By focusing on machine learning approaches like transfer learning, knowledge distillation, reinforcement learning, deep sequential models, and deep ensemble learning, this thesis endeavors to establish efficient and accurate anomaly detection systems specific for edge cloud environments.The results demonstrate the improvements achieved by employing machine learning methods for anomaly detection in edge clouds. Extensive testing and evaluation in real-world edge environments show how machine learning-driven anomaly detection systems improve identification of anomalies in edge clouds. The results highlight the capability of these methods to achieve a reasonable trade-off between accuracy and computational efficiency. These findings illustrate how machine learning-based anomaly detection approaches contribute to building resilient and secure edge-based systems.
|
|
2. |
- Ibidunmoye, Olumuyiwa, 1983-
(författare)
-
Performance problem diagnosis in cloud infrastructures
- 2016
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Cloud datacenters comprise hundreds or thousands of disparate application services, each having stringent performance and availability requirements, sharing a finite set of heterogeneous hardware and software resources. The implication of such complex environment is that the occurrence of performance problems, such as slow application response and unplanned downtimes, has become a norm rather than exception resulting in decreased revenue, damaged reputation, and huge human-effort in diagnosis. Though causes can be as varied as application issues (e.g. bugs), machine-level failures (e.g. faulty server), and operator errors (e.g. mis-configurations), recent studies have attributed capacity-related issues, such as resource shortage and contention, as the cause of most performance problems on the Internet today. As cloud datacenters become increasingly autonomous there is need for automated performance diagnosis systems that can adapt their operation to reflect the changing workload and topology in the infrastructure. In particular, such systems should be able to detect anomalous performance events, uncover manifestations of capacity bottlenecks, localize actual root-cause(s), and possibly suggest or actuate corrections.This thesis investigates approaches for diagnosing performance problems in cloud infrastructures. We present the outcome of an extensive survey of existing research contributions addressing performance diagnosis in diverse systems domains. We also present models and algorithms for detecting anomalies in real-time application performance and identification of anomalous datacenter resources based on operational metrics and spatial dependency across datacenter components. Empirical evaluations of our approaches shows how they can be used to improve end-user experience, service assurance and support root-cause analysis.
|
|
3. |
- Karlsson, Lars, 1982-
(författare)
-
Scheduling of parallel matrix computations and data layout conversion for HPC and Multi-Core Architectures
- 2011
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Dense linear algebra represents fundamental building blocks in many computational science and engineering applications. The dense linear algebra algorithms must be numerically stable, robust, and reliable in order to be usable as black-box solvers by expert as well as non-expert users. The algorithms also need to scale and run efficiently on massively parallel computers with multi-core nodes. Developing high-performance algorithms for dense matrix computations is a challenging task, especially since the widespread adoption of multi-core architectures. Cache reuse is an even more critical issue on multi-core processors than on uni-core processors due to their larger computational power and more complex memory hierarchies. Blocked matrix storage formats, in which blocks of the matrix are stored contiguously, and blocked algorithms, in which the algorithms exhibit large amounts of cache reuse, remain key techniques in the effort to approach the theoretical peak performance. In Paper I, we present a packed and distributed Cholesky factorization algorithm based on a new blocked and packed matrix storage format. High performance node computations are obtained as a result of the blocked storage format, and the use of look-ahead leads to improved parallel efficiency. In Paper II and Paper III, we study the problem of in-place matrix transposition in general and in-place matrix storage format conversion in particular. We present and evaluate new high-performance parallel algorithms for in-place conversion between the standard column-major and row-major formats and the four standard blocked matrix storage formats. Another critical issue, besides cache reuse, is that of efficient scheduling of computational tasks. Many weakly scalable parallel algorithms are efficient only when the problem size per processor is relatively large. A current research trend focuses on developing parallel algorithms which are more strongly scalable and hence more efficient also for smaller problems. In Paper IV, we present a framework for dynamic node-scheduling of two-sided matrix computations and demonstrate that by using priority-based scheduling one can obtain an efficient scheduling of a QR sweep. In Paper V and Paper VI, we present a blocked implementation of two-stage Hessenberg reduction targeting multi-core architectures. The main contributions of Paper V are in the blocking and scheduling of the second stage. Specifically, we show that the concept of look-ahead can be applied also to this two-sided factorization, and we propose an adaptive load-balancing technique that allow us to schedule the operations effectively.
|
|
4. |
- Lakew, Ewnetu Bayuh
(författare)
-
Managing Resource Usage and Allocations in Multi-Cluster Clouds
- 2013
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- The emergence of large-scale Internet services has fueled a trend toward large-scale systems composed of geographically distributed clusters. Managing resource allocations and resource usage is an important task for such services.Resource allocations and resource usage management mechanisms for services running across clusters play vital roles in the performance of the entire system, economical sustainability of the provider, and level of customers satisfaction provided by the system. However, when providing the utmost customer satisfaction the service provider ought to make sure not to over-commit resources beyond the agreed limit between the customer and the provider. Moreover, statistics of resources consumed by different services should be monitored and collected using an efficient mechanism with minimal overhead and interference on the system and the services. Thus, resource usage collection and allocations mechanisms should impose economical constraints to both sides, the customer and the cloud provider.This thesis focuses on decentralized resource allocation and resource usage management for services running in multi cluster environments. Theoretical as well as experimental results indicate that our proposed approaches provide efficient management of resources for services running in a large-scale geographically distributed systems.
|
|
5. |
- Li, Wubin, 1983-
(författare)
-
Algorithms and Systems for Virtual Machine Scheduling in Cloud Infrastructures
- 2014
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- With the emergence of cloud computing, computing resources (i.e., networks, servers, storage, applications, etc.) are provisioned as metered on-demand services over net- works, and can be rapidly allocated and released with minimal management effort. In the cloud computing paradigm, the virtual machine (VM) is one of the most com- monly used resource units in which business services are encapsulated. VM schedul- ing optimization, i.e., finding optimal placement schemes for VMs and reconfigu- rations according to the changing conditions, becomes challenging issues for cloud infrastructure providers and their customers.The thesis investigates the VM scheduling problem in two scenarios: (i) single- cloud environments where VMs are scheduled within a cloud aiming at improving criteria such as load balancing, carbon footprint, utilization, and revenue, and (ii) multi-cloud scenarios where a cloud user (which could be the owner of the VMs or a cloud infrastructure provider) schedules VMs across multiple cloud providers, target- ing optimization for investment cost, service availability, etc. For single-cloud scenar- ios, taking load balancing as the objective, an approach to optimal VM placement for predictable and time-constrained peak loads is presented. In addition, we also present a set of heuristic methods based on fundamental management actions (namely, sus- pend and resume physical machines, VM migration, and suspend and resume VMs), continuously optimizing the profit for the cloud infrastructure provider regardless of the predictability of the workload. For multi-cloud scenarios, we identify key re- quirements for service deployment in a range of common cloud scenarios (including private clouds, bursted clouds, federated clouds, multi-clouds, and cloud brokering), and present a general architecture to meet these requirements. Based on this architec- ture, a set of placement algorithms tuned for cost optimization under dynamic pricing schemes are evaluated. By explicitly specifying service structure, component relation- ships, and placement constraints, a mechanism is introduced to enable service owners the ability to influence placement. In addition, we also study how dynamic cloud scheduling using VM migration can be modeled using a linear integer programming approach.The primary contribution of this thesis is the development and evaluation of al- gorithms (ranging from combinatorial optimization formulations to simple heuristic algorithms) for VM scheduling in cloud infrastructures. In addition to scientific pub- lications, this work also contributes software tools (in the OPTIMIS project funded by the European Commissions Seventh Framework Programme) that demonstrate the feasibility and characteristics of the approaches presented.
|
|
6. |
- Mehta, Amardeep, 1985-
(författare)
-
Resource allocation for Mobile Edge Clouds
- 2018
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Recent advances in Internet technologies have led to the proliferation of new distributed applications in the transportation, healthcare, mining, security, and entertainment sectors. The emerging applications have characteristics such as being bandwidth-hungry, latency-critical, and applications with a user population contained within a limited geographical area, and require high availability, low jitter, and security.One way of addressing the challenges arising because of these emerging applications, is to move the computing capabilities closer to the end-users, at the logical edge of a network, in order to improve the performance, operating cost, and reliability of applications and services. These distributed new resources and software stacks, situated on the path between today's centralized data centers and devices in close proximity to the last mile network, are known as Mobile Edge Clouds (MECs). The distributed MECs provides new opportunities for the management of compute resources and the allocation of applications to those resources in order to minimize the overall cost of application deployment while satisfying end-user demands in terms of application performance.However, these opportunities also present three significant challenges. The first challenge is where and how much computing resources to deploy along the path between today's centralized data centers and devices for cost-optimal operations. The second challenge is where and how much resources should be allocated to which applications to meet the applications' performance requirements while minimizing operational costs. The third challenge is how to provide a framework for application deployment on resource-constrained IoT devices in heterogeneous environments. This thesis addresses the above challenges by proposing several models, algorithms, and simulation and software frameworks. In the first part, we investigate methods for early detection of short-lived and significant increase in demand for computing resources (also called spikes) which may cause significant degradation in the performance of a distributed application. We make use of adaptive signal processing techniques for early detection of spikes. We then consider trade-offs between parameters such as the time taken to detect a spike and the number of false spikes that are detected. In the second part, we study the resource planning problem where we study the cost benefits of adding new compute resources based on performance requirements for emerging applications. In the third part, we study the problem of allocating resources to applications by formulating as an optimization problem, where the objective is to minimize overall operational cost while meeting the performance targets of applications. We also propose a hierarchical scheduling framework and policies for allocating resources to applications based on performance metrics of both applications and compute resources. In the last part, we propose a framework, Calvin Constrained, for resource-constrained devices, which is an extension of the Calvin framework and supports a limited but essential subset of the features of the reference framework taking into account the limited memory and processing power of the resource-constrained IoT devices.
|
|
7. |
- Rahmanian, Ali, 1989-
(författare)
-
Edge orchestration for latency-sensitive applications
- 2024
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- The emerging edge computing infrastructure provides distributed and heterogeneous resources closer to where data is generated and where end-users are located, thereby significantly reducing latency. With the recent advances in telecommunication systems, software architecture, and machine learning, there is a noticeable increase in applications that require processing times within tight latency constraints, i.e. latency-sensitive applications. For instance, numerous video analytics applications, such as traffic control systems, necessitate real-time processing capabilities. Orchestrating such applications at the edge offers numerous advantages, including lower latency, optimized bandwidth utilization, and enhanced scalability. However, despite its potential, effectively managing such latency-sensitive applications at the edge poses several challenges such as constrained compute resources, which holds back the full promise of edge computing.This thesis proposes approaches to efficiently deploy latency-sensitive applications on the edge infrastructure. It partly addresses general applications with microservice architectures and party addresses the increasingly more important video analytics applications for the edge. To do so, this thesis proposes various application- and system-level solutions aiming to efficiently utilize constrained compute capacity on the edge while meeting prescribed latency constraints. These solutions primarily focus on effective resource management approaches and optimizing incoming workload inputs, considering the constrained compute capacity of edge resources. Additionally, the thesis explores the synergy effects of employing both application- and system-level resource optimization approaches together.The results demonstrate the effectiveness of the proposed solutions in enhancing the utilization of edge resources for latency-sensitive applications while adhering to application constraints. The proposed resource management solutions, alongside application-level optimization techniques, significantly improve resource efficiency while satisfying application requirements. Our results show that our solutions for microservice architectures significantly improve end-to-end latency by up to 800% while minimizing edge resource usage. Additionally, the results indicate that our application- and system-level optimizations for orchestrating edge resources for video analytics applications can increase the overall throughput by up to 60%.
|
|
8. |
- Sundqvist, Tobias, 1976-
(författare)
-
Machine learning-based diagnostics and observability in mobile networks
- 2023
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- To meet the high-performance and reliability demands of 5G, the Radio Access Network (RAN) is moving to a cloud-native architecture. The new microservice architecture promises increased operational efficiency and a shorter time-to-market, but it also comes with a price. The new distributed and virtualized architecture is far more complex than ever before, and with the increasing number of features it brings, troubleshooting becomes more difficult. So far, RAN troubleshooters have relied on their expertise to analyze systems manually, but the ever-growing data and increased complexity make it challenging to grasp system behavior.This thesis contributes threefold, where the proposed machine learning and statistical methods help RAN troubleshooters find deviations in system logs, identify the root cause of these deviations, and improve the system's observability. These methods learn the application's behavior from the system logs events and can identify behavior deviations from many different aspects. The thesis also demonstrates how observability can be improved by using a new software instrumentation guideline. The guideline enables the tracking of systemized procedures and enhances system understanding. The purpose of the guideline is to make RAN developers aware that machine learning can utilize debug information and help their troubleshooting process. To familiarize the reader with the research area, the challenges, and methods that can be used to detect anomalies, perform root cause analysis and observe RAN system behavior. The proposed research methods are integrated and tested in an advanced 5G test bed to evaluate the methods' accuracy, speed, system impact, and implementation cost.The results demonstrate the advantage of using machine learning and statistical methods when troubleshooting the behavior of RAN. Machine learning methods, similar to those presented in this thesis, may help those who troubleshoot RAN and accelerate the development of 5G. The thesis ends with presenting potential research areas where this research could be further developed and applied, both in RAN and other systems.
|
|
9. |
- Gonzalo P., Rodrigo, 1980-
(författare)
-
HPC scheduling in a brave new world
- 2017
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Many breakthroughs in scientific and industrial research are supported by simulations and calculations performed on high performance computing (HPC) systems. These systems typically consist of uniform, largely parallel compute resources and high bandwidth concurrent file systems interconnected by low latency synchronous networks. HPC systems are managed by batch schedulers that order the execution of application jobs to maximize utilization while steering turnaround time. In the past, demands for greater capacity were met by building more powerful systems with more compute nodes, greater transistor densities, and higher processor operating frequencies. Unfortunately, the scope for further increases in processor frequency is restricted by the limitations of semiconductor technology. Instead, parallelism within processors and in numbers of compute nodes is increasing, while the capacity of single processing units remains unchanged. In addition, HPC systems’ memory and I/O hierarchies are becoming deeper and more complex to keep up with the systems’ processing power. HPC applications are also changing: the need to analyze large data sets and simulation results is increasing the importance of data processing and data-intensive applications. Moreover, composition of applications through workflows within HPC centers is becoming increasingly important. This thesis addresses the HPC scheduling challenges created by such new systems and applications. It begins with a detailed analysis of the evolution of the workloads of three reference HPC systems at the National Energy Research Supercomputing Center (NERSC), with a focus on job heterogeneity and scheduler performance. This is followed by an analysis and improvement of a fairshare prioritization mechanism for HPC schedulers. The thesis then surveys the current state of the art and expected near-future developments in HPC hardware and applications, and identifies unaddressed scheduling challenges that they will introduce. These challenges include application diversity and issues with workflow scheduling or the scheduling of I/O resources to support applications. Next, a cloud-inspired HPC scheduling model is presented that can accommodate application diversity, takes advantage of malleable applications, and enables short wait times for applications. Finally, to support ongoing scheduling research, an open source scheduling simulation framework is proposed that allows new scheduling algorithms to be implemented and evaluated in a production scheduler using workloads modeled on those of a real system. The thesis concludes with the presentation of a workflow scheduling algorithm to minimize workflows’ turnaround time without over-allocating resources.
|
|
10. |
- Ibidunmoye, Olumuyiwa, 1983-
(författare)
-
Performance anomaly detection and resolution for autonomous clouds
- 2017
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Fundamental properties of cloud computing such as resource sharing and on-demand self-servicing is driving a growing adoption of the cloud for hosting both legacy and new application services. A consequence of this growth is that the increasing scale and complexity of the underlying cloud infrastructure as well as the fluctuating service workloads is inducing performance incidents at a higher frequency than ever before with far-reaching impact on revenue, reliability, and reputation. Hence, effectively managing performance incidents with emphasis on timely detection, diagnosis and resolution has thus become a necessity rather than luxury. While other aspects of cloud management such as monitoring and resource management are experiencing greater automation, automated management of performance incidents remains a major concern.Given the volume of operational data produced by cloud datacenters and services, this thesis focus on how data analytics techniques can be used in the aspect of cloud performance management. In particular, this work investigates techniques and models for automated performance anomaly detection and prevention in cloud environments. To familiarize with developments in the research area, we present the outcome of an extensive survey of existing research contributions addressing various aspects of performance problem management in diverse systems domains. We discuss the design and evaluation of analytics models and algorithms for detecting performance anomalies in real-time behaviour of cloud datacenter resources and hosted services at different resolutions. We also discuss the design of a semi-supervised machine learning approach for mitigating performance degradation by actively driving quality of service from undesirable states to a desired target state via incremental capacity optimization. The research methods used in this thesis include experiments on real virtualized testbeds to evaluate aspects of proposed techniques while other aspects are evaluated using performance traces from real-world datacenters.Insights and outcomes from this thesis can be used by both cloud and service operators to enhance the automation of performance problem detection, diagnosis and resolution. They also have the potential to spur further research in the area while being applicable in related domains such as Internet of Things (IoT), industrial sensors as well as in edge and mobile clouds.
|
|